INTRODUCTION

Primary liver cancer (PLC) includes a diverse group of malignant tumors such as hepatocellular carcinoma (HCC), cholangiocarcinoma, and combined hepatocellular-intrahepatic cholangiocarcinoma. All of these tumors have distinct histopathological and molecular profiles. Among these, HCC alone, which originates from the malignant transformation of hepatocytes, represents 80-90% of PLC cases.¹ The pathogenesis involves an interaction between environmental and genetic factors. Several factors contribute to the development of HCC, such as hepatitis C virus/hepatitis B virus (HCV/HBV) infection, alcohol abuse, non-alcoholic fatty liver disease, and contact with environmental carcinogens (e.g., aflatoxin B1). Less common factors include α1-antitrypsin deficiency, hepatolenticular degeneration or Wilson’s disease, erythropoietic protoporphyria, and autoimmune hepatitis.² The precise molecular mechanisms of HCC are complicated and heterogeneous, impeding clinical diagnosis and treatment. It poses a significant health threat to patients because of its high relapse rate and poor response to current therapies. For instance, treatment with tyrosine kinase inhibitors (e.g., lenvatinib and sorafenib) or immune checkpoint inhibitors (e.g., atezolizumab combined with bevacizumab) offers limited clinical benefits.³ The presence of intertumoral and intratumoral heterogeneity is an important factor that contributes to its limited efficacy. This heterogeneity is driven by complex interactions between cancer cells and the surrounding tumor microenvironment (TME), comprising stromal, immune, mesenchymal, and endothelial cells, leading to the emergence of various HCC sub-clusters that remodel the tumor ecosystem to favor HCC growth by manipulating neighboring cells (Fig. 1).⁴ Although several studies have shown that liver cancer comprises a variety of cells, the effect of the differential nature of cells on the clinical outcome of HCC remains unclear.^5,6 Therefore, an in-depth or detailed characterization of the tumor ecosystem involved in HCC initiation, progression, and therapeutic resistance is essential for developing more effective diagnostic and treatment strategies.

Conventional sequencing methods, such as whole-exome sequencing, whole-transcriptome sequencing, and epigenomic profiling, have limited efficacy in understanding the complexity of tumor heterogeneity because they rely on bulk tissue analysis, which has the limitation of identifying signals from rare cell populations. In contrast, single-cell sequencing is a powerful technique that can overcome this challenge by permitting detailed analysis of the interactions between tumor cells and various cell types within the TME. It also provides valuable information regarding the functional dynamics and composition of the tumor ecosystem. Unimodal approaches to single-cell sequencing study single cells at the genome, transcriptome, epigenome, or proteome level independently. In contrast, single-cell multi-omics integrates multiple unimodal datasets, offering a comprehensive view of the complex relationships between genotypes and phenotypes in patients with HCC. Recent technological advancements in single-cell and spatial omics have provided a more refined and high-resolution understanding of tumor biology, offering new opportunities for precision oncology. For instance, a novel spatial transcriptome platform, Stereo-seq, was used to identify a 500-μm-wide invasive zone centered around the tumor border in patients with HCC.⁷ The same method was used to assess heterogeneity of cancer-associated fibroblasts (CAFs) and identified two subsets: CAF-fibroblast activation protein (FAP) and CAF-C7. The enrichment of these subtypes strongly correlated with worse outcomes.⁸ These studies suggest that spatial omics, when integrated with single-cell approaches, can help identify the real cause or resistance behavior of a disease. This may accelerate the discovery of effective treatments for this complex disease.

In this review, we provide an overview of the main single-cell sequencing techniques and summarize the impact of these technologies on understanding the heterogeneous behavior of liver cancer cells and the immune cell microenvironment.

SINGLE-CELL OMICS TECHNOLOGIES AND HCC

In contrast to conventional sequencing, single-cell sequencing has dramatically improved accuracy and efficiency. This technique utilizes several methods, including single-cell DNA sequencing (scDNA-seq) for DNA, single-cell RNA sequencing (scRNA-seq) for RNA, cytometry by time of flight (CyTOF) for cell surface proteins, and single-cell assay for transposase-accessible chromatin using sequencing (scATAC-seq) for chromatin accessibility. A single cell contains DNA or RNA in picograms. Therefore, whole-genome amplification (WGA) is required to obtain sufficient nucleic acid material. WGA has been previously performed using degenerate oligonucleotide-primed polymerase chain reaction (PCR), interspersed repetitive sequence PCR, primer extension pre-amplification PCR, and linker-adapter PCR. These WGA techniques have been replaced in modern scDNA-seq with multiple displacement amplification (MDA), multiple annealing and looping-based amplification cycles (MALBAC), and linear amplification via transposon insertion (LIANTI). In MDA, high-fidelity Phi29 polymerase is used for genome amplification to achieve high coverage of genes and low error rates compared with PCR. In MALBAC, primers have a 27 bp common sequence at the 5'-end and an 8 bp random sequence at the 3'-end, which enables copying of genomic DNA. After the first round of amplification, the amplified product contains complementary sequences at both 5'- and 3'-ends and thus forms a self-loop. This loop prevents further amplification of this strand, thereby preventing amplification bias and increasing genomic coverage.⁹ In LIANTI, transposases (such as Tn5) are used to insert primers or adaptors into genomic DNA at many sites. This process is called tagmentation. Only one primer is used to perform linear amplification. Because no reverse primer is used, each strand is extended once per cycle, thereby preventing exponential doubling. Similarly, the initial step in an scRNA-seq experiment is the isolation of single cells, which can be performed using several methods such as fluorescence-activated cell sorting, chip-based or microdroplet-based microfluidic technologies, or laser microdissection.¹⁰ Following this, scRNA-seq libraries are generated by cell lysis, reverse transcription, second-strand synthesis of complementary DNA (cDNA), and amplification, followed by deep sequencing.

scRNA-seq datasets are usually visualized using a t-distributed stochastic neighbor embedding (t-SNE) map. In this map, cells are clustered based on their transcriptome similarity, and single or multiple genes can be visualized separately on individual t-SNE maps.¹⁰ Several methods are used for scRNA-seq, such as STRT-seq, SMART-seq, CEL-seq, DROP-seq, and 10X Chromium Genomics (Fig. 2). However, these techniques differ in terms of the specific steps and workflows involved in scRNA-seq. These scRNA-seq methods have various limitations; therefore, researchers choose platforms that are best suited to their study objectives. For example, the SMART-seq platform has been used to reveal a high-resolution dynamic immune landscape in HCC. In this study, we demonstrated that LAMP3+ dendritic cells (DCs) express diverse immune-relevant ligands. These DCs can migrate from tumors to the hepatic lymph nodes and regulate lymphocytes. The study also characterized tumor-associated macrophage states linked to poor prognosis, providing one of the earliest comprehensive single-cell maps of the HCC immune microenvironment.¹¹ In another study, the SMART-seq2 sequencing method was used on heterogeneous subclones in HCC, including five HCC and two hepatocyte subclones. The study found that MLX-interacting proteins-like (MLXIPL) were commonly upregulated in HCC and associated with poor survival. These results identified that MLXIPL in HCC is a promising therapeutic target.¹² Similarly, different regions of HCC and intrahepatic cholangiocarcinoma were analyzed using droplet-based 5'-scRNA-seq to determine the spatial distribution of tumor cells and the TME. This study observed that the diversity between HCC and intrahepatic cholangiocarcinoma was higher than that within the tumor, regardless of tumor size and the location of sampling tissues.¹³ While scRNA-seq reveals the transcriptional heterogeneity of tumor and stromal cells in HCC, single-cell epigenomic approaches are critical for understanding the regulatory mechanisms that drive these expression patterns. For epigenomes, several methods are used, including RRBS, WGBS, CGI-seq, ATAC-seq, and ChIP-seq. The mechanisms, advantages, and disadvantages of these scRNA-seq and single-cell epigenomic methods are described in Table 1^9,14-22 and Table 2,^9,23-29 respectively.

The most commonly used tools for scRNA-seq analysis include the R package Seurat, Python package Scanpy, SeuratDisk for interoperability between Seurat and Scanpy, scDIOR for rapidly transforming data between R and Python, and CellBridge.³⁰ In addition to these tools, several other tools are also used, including FastQC, CellRanger, Trimmomatic, STAR for preprocessing, doubletFinder, Scrublet for quality control, Seurat, Gini-Clust for feature selection, t-SNE and uniform manifold approximation and projection (UMAP) for dimensionality reduction and visualization, and CellAssign for annotation.³¹ EpiScanpy is a versatile toolkit used for scDNA methylation and scATAC-seq data.³²

MULTI-OMICS AND HCC

Unimodal methods of single-cell sequencing are combined into multi-omics approaches that enable the acquisition of multiple types of information from the same single cell simultaneously. Multi-omics approaches comprise several methods.

CITE-seq

This method is used to study the transcriptome and surface proteins. In this technique, oligo-tagged antibodies with poly(A) tails are used; therefore, in contrast to flow cytometry, there is no upper limit to the number of antibodies that can be used to identify cell surface proteins. In addition, cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq) can be adapted for RNA interference, CRISPR, and other gene-editing techniques. Bioinformatics tools such as the CITE-seq Python package, CellRanger,³³ Seurat, total VI, CiteFuse, and LinQ-View provide frameworks for CITE-seq data analysis, including dimensionality reduction, cell clustering, cell type identification, and differential expression of RNA and cell surface proteins.³⁴ The data can also be analyzed using SeqGeq or FlowJo software (Ashland, OR, USA).³⁵ However, CITE-seq has several limitations, such as low throughput, poor resolution, no detection of intracellular proteins, loss of information about the location of cells, availability of antibodies, and optimization of the assay.³⁴

This method was recently used to identify six liver cancer stem cell markers (CD44, EPCAM, CD133, CD24, CD90, and CD54) in five HCC cell lines. These markers characterize the epithelial or mesenchymal state of these cells. The study also identified a robust gene panel representing the hypoxia signature in specific high-yield glycolysis clusters and found a correlation with HCC prognosis.³⁶

Techniques enabling the parallel analysis of genomic DNA and polyadenylated RNA within individual cells

DR-seq

In this preamplification, genomic DNA (gDNA) and RNA are amplified simultaneously to minimize nucleic acid loss. Following this, the reaction is divided into two tubes to complete the gDNA and RNA sequencing libraries separately.³⁷ However, the analysis and interpretation of such data are complicated by the presence of contaminating RNA sequences within the gDNA sequencing dataset.

scONE-seq

In this technique, gDNA and RNA are differentially barcoded; therefore, sequencing of RNA and DNA is possible in a single reaction. It is particularly useful for frozen samples because DNA and RNA are not physically separated.³⁸ However, re-sequencing of genomic material at greater depths is not possible, and the method lacks flexibility for selecting subsets of gDNA or RNA for sequencing.

G&T-seq and DNTR-seq

In these methods, RNA and gDNA are physically separated; therefore, these methods are flexible and scalable.³⁹ However, gDNA sequencing in genome and transcriptome sequencing (G&T-seq) necessitates WGA, making it expensive and time-consuming.

Gtag&T-seq

In this method, WGA is not performed to sequence the gDNA and RNA of single cells, thus enhancing throughput while minimizing coverage bias, amplification noise, and cost.⁴⁰

In the field of liver cancer, no widely recognized published study has applied DR-seq, scONE-seq, G&T-seq, DNTR-seq, or Gtag&T-seq. There could be several reasons for this, including technical complexity, preference for alternative methods such as single-cell triple omics sequencing (cTrio-seq), single-nucleus chromatin accessibility and mRNA expression sequencing (SNARE-seq), 10X Multiome, or spatial transcriptomics, or a focus on transcriptome profiling using scRNA-seq alone.

SUPeR-seq

This method is used to identify and characterize super-enhancers (SEs) in various cancers, focusing on both full-length and small RNA sequencing. SEs contain approximately 10-fold higher density of transcription factors, making them superior to typical enhancers.⁴¹ SE-associated genes and long non-coding RNAs (lncRNAs) play an important role in the progression and development of HCC. For example, SE-targeted genes (SPHK1, SPIDR, AJUBA, QK1, and SIRT7), microRNAs (miR-9), and lncRNAs (HCCL5, LINC1089, lncRNA-DAW, and HSAL-3) play crucial roles in HCC progression.⁴¹ Recently, a study has shown that E2F1 combined with LINC01004, a novel SE-associated lncRNA, is upregulated in liver cancer tissues and is associated with cell proliferation and metastasis with poor patient prognosis.⁴² Therefore, single-cell universal poly(A)-independent RNA sequencing (SUPeR-seq) could be instrumental in uncovering the role of SE-associated target genes and lncRNAs in HCC; however, to date, its use in HCC has not been reported.

scTrio-seq

This single-cell triple omics sequencing technique can be used to analyze genomic copy number variations (CNVs), DNA methylomes, and transcriptomes simultaneously in a single cell. The unique and specific arrangement of DNA methylation abnormalities has been correlated with the diagnosis and prognosis of HCC.² This technique was applied to 25 single HCC cancer cells, and two subpopulations were identified within these cells based on CNVs, DNA methylome, or transcriptome of individual cells.⁴³ Subpopulations expressing high levels of cell invasion biomarkers tend to escape immune surveillance. Thus, filtering and capturing different subpopulations of tumor cells is an important topic for future cancer research.

SNARE-seq

This method simultaneously measures chromatin accessibility and gene expression.⁴⁴ Briefly, in this technique, accessible sites are captured by Tn5 transposase in permeabilized nuclei to permit DNA barcode tagging together with the mRNA molecules from the same cells.⁴⁵ Similarly, ultra-high-throughput joint analysis of the transcriptome and accessible chromatin can be performed using paired-seq to study millions of individual cells simultaneously. This method simultaneously tags open chromatin fragments generated by Tn5 transposases and cDNA molecules generated from reverse transcription of RNA in millions of cells. This technique was applied to Hep-G2 cells, in which reads from the DNA library were enriched around transcription start sites, whereas reads from the RNA library were enriched upstream of transcription termination sites.⁴⁶

scRNA-seq + scTCR-seq

This technique combines gene expression and T-cell receptor (TCR) sequences from the same immune cells3 and has been applied to HCC. For example, tumors from patients with HCC treated with atezolizumab plus bevacizumab that presented with durable responses were enriched in PD-L1⁺ CXCL10⁺ macrophages and, based on cell-cell interaction analysis, expressed high levels of CXCL9/10/11 and were predicted to attract peripheral CXCR3⁺ CD8⁺ effector memory T cells (CD8 TEM) into the tumor. These data suggest that CD8 TEM cells preferentially differentiate into clonally expanded PD-1^- CD45RA⁺ effector-memory CD8⁺ T cells (CD8 TEMRA) with pronounced cytotoxicity. However, contradictory results were obtained in non-responders.³³ Another study analyzed the mutational profiles and evolutionary trajectories of paired primary and recurrent HCC samples and identified differential gene expression in de novo versus true recurrence. The abundance of KLRB1⁺ CD8⁺ T cells and growth differentiation factor (GDF) 15 in DCs was high in truly recurrent HCCs with low cytotoxicity. High GDF4 levels in DCs dampen antigen presentation and inhibit antitumor immunity in recurrent lesions. Consistent with these findings, a phase 2 trial of neoadjuvant anti-PD-1 immunotherapy showed better responses in patients with de novo recurrent HCC.⁴⁷

SINGLE-CELL OMICS AND MULTI-OMICS IN UNRAVELING HETEROGENEITY IN HCC

Tumor heterogeneity incorporates various subpopulations of cells, either within a tumor (intratumoral heterogeneity) or between different tumors or nodes in the same patient (intertumoral heterogeneity). The biological behavior of these cells differs owing to their different genetic and morphological characteristics, including size, shape, and structure. Heterogeneity also exists before and after tumor treatment, which is known as spatiotemporal heterogeneity. For example, berberine treatment alters the tumor immune landscape in mice by decreasing the proportion of TCR-bhiPD-1hiCD69⁺ CD27⁺ effector CD8⁺ T lymphocytes and increasing the proportion of Ly6ChiTCRb⁺ CD69⁺ CD27⁺ CD62L⁺ central memory CD8⁺ T lymphocytes.⁴⁸ Heterogeneity is commonly observed in HCC and is thought to result from differences in patient genetics and TME. The diverse pathways involved in HCC contribute to its clinical complexity and are crucial factors in challenges, such as treatment resistance, tumor dormancy, and relapse. Several factors, such as genomic mutations, epigenomic changes, and the TME, contribute to the clonal evolution of cancer, resulting in heterogeneity at the genomic, molecular, and functional levels.

The TME, particularly immune cells and CAFs, plays a key role in shaping HCC heterogeneity and treatment resistance. CAFs promote tumorigenesis by secreting cytokines, chemokines, and growth factors. Notably, the liver is in direct contact with bacterial products or other inflammatory stimuli and has the highest number of immune cells, such as Kupffer cells. Hence, it prevents excessive immune responses to maintain tissue health. Dysregulation of the immunological network due to prolonged inflammation can lead to HCC. Therefore, the interactions between hepatic immune cells and malignant hepatocytes should be studied to improve understanding of HCC and develop new treatment regimens. scRNA-seq studies have revealed CAF subtypes, such as POSTN⁺,^49,50 CD36⁺,⁵¹ and vascular endothelial growth factor A (VEGFA)⁺,⁵² that promote immunosuppression, recruitment of myeloid-derived suppressor cells, and angiogenesis, respectively. These findings underscore the importance of mapping the immune TME landscape for the development of targeted therapies. Recent scRNA-seq of CAF populations in HCC further revealed the immunosuppressive role of immunoglobulin A (IgA) signaling in the TME. The PD-L1 level in IgA-treated CAFs was increased, and the amount of fibroblast activation protein was elevated, which suppressed the cytotoxicity of CD8⁺ T cells through PD-1-induced interactions. This study suggests that intrahepatic IgA induces polarization of HCCCAFs into more malignant matrix phenotypes and attenuates the function of cytotoxic T cells. These findings highlight their potential roles in tumor progression and immune suppression.⁵³

Single-cell sequencing technologies have become essential for uncovering cellular and molecular landscapes.⁵² Alvarez et al.⁵⁴ used single-cell analysis in human liver single nucleus. They identified a proliferating cell type carrying TP53 and RB1 somatic mutations that are enriched in HCC tumors and associated with decreased overall survival. The six datasets of scRNA-seq identified five clusters and found that the aggressive HCC subtype is composed of a combination of cluster 1 characterized by high expression of p53 and myc with cluster 3, 4, or 5. Cluster 1 elucidated T cell depletion as an immune resistance mechanism. This study deepens our understanding of the molecular mechanisms and TME in aggressive HCC.⁵⁵ To move toward clinical application, an 11-gene prognostic signature was derived from the differentially expressed genes, distinguishing tumor from adjacent non-tumor cells using an scRNA-seq approach. This signature stratifies patients with HCC into two risk groups based on markedly different molecular profiles. The study also identified 14 candidate compounds that may improve outcomes in aggressive HCC in the high-risk group.⁵⁶ These studies collectively highlight the power of single-cell sequencing technologies in identifying distinct tumor subpopulations, prognostic gene signatures, and potential therapeutic targets.

However, scRNA-seq technology still has limitations due to the loss of spatial and morphological information after tissue dissociation, which makes it difficult to determine the spatial architecture of tumors. Methods such as multiplexed error-robust fluorescence in situ hybridization and sequential fluorescence in situ hybridization can capture spatial information; however, they are limited to detecting only a small number of predefined target genes simultaneously.⁵⁷ The newly developed spatial transcriptomics technology can overcome these limitations. For example, the spatial transcriptome architecture of seven PLCs characterized the tertiary lymphoid structure composition, stromal and immune cell distribution, cancer stem cell niche diversity, and tumor cluster interactions. In brief, the study revealed that reciprocal ligand-receptor interactions occurring along the 100-μm cluster boundary are essential for preserving the structural organization within the tumor. PROM1⁺ CD47⁺ cancer stem cell niches contribute to the remodeling of the TME and tumor metastasis.⁵⁷ These findings provide a deeper understanding of the complex ecosystem of liver cancer and can therefore improve individualized cancer prevention and drug discovery.

scRNA-seq and spatial transcriptomics are inadequate for epigenetics, proteomics, and metabolomics. Thus, multi-omics with bulk tissue-resolution profiling has the potential to address such deficiencies. For example, using a combination of scRNA-seq, spatial transcriptomics, and bulk multi-omics, Zhou et al.⁵⁸ revealed that intrahepatic cholangiocarcinoma was the primary source of CAFs, whereas HCC exhibited disrupted metabolism and greater individual heterogeneity of T cells. Similarly, in another study, single-cell transcriptomic, proteomic, and chromatin accessibility data were combined to identify the correlation between heterogeneity and metastatic potential in five HCC cell lines. They identified a rare hypoxic subtype with a high capacity for glycolysis and increased metastatic potential. Additionally, they identified a robust 14-gene panel representing the hypoxia signature that could serve as a prognostic index.³⁶ In a recent study on HCC, single-cell spatial transcriptomics and bulk multiomics analyses were used to investigate the heterogeneous behavior of HCC and its ecosystems. They identified MI types of tumor-associated macrophages that perturb metabolism, antigen presentation, and immune-killing abilities. In addition, they identified that cirrhotic and HCC lesions in an individual patient shared a common origin and then evolved in parallel; each lesion adopted distinct immune reprogramming strategies to adapt to its local microenvironment.⁴ This dataset revealed the dynamic heterogeneity of HCC across spatiotemporal dimensions. Patients with β-catenin (encoded by CTNNB1)-mutated HCC demonstrate heterogeneous responses to first-line immune checkpoint inhibitors. Precision medicine-based treatments are not available for this class. Analysis of bulk and spatial transcriptomic data showed that the mutated β-catenin gene signature overlapped with the Hoshida S3, Boyault G5/G6, and Chiang CTNNB1 subclasses in The Cancer Genome Atlas and HCC spatial datasets. Therefore, this signature may aid in patient stratification to guide precision medicine therapeutics for the CTNNB1-mutated HCC subclass.⁵⁹ These studies highlight the importance of multi-omics technologies in addressing the spatial, functional, and evolutionary complexities of HCC (Fig. 3).

Given the pressing need for novel liver cancer therapies, scRNAseq not only uncovers potential drug targets but also enables finer tumor subclassification to better tailor individual treatments. An initial scRNA-seq-based classification system has been proposed. However, it awaits prospective validation before adoption in routine clinical practice. To date, only a handful of single-cell sequencing studies (Table 3) and clinical trials on HCC have investigated therapies based on single-cell analyses, highlighting the gap between their discovery and application (Table 4). Combining scRNA-seq with spatial transcriptomics will deepen our insight into cellular and molecular crosstalk within the TME, driving the discovery of new targets in hepatobiliary cancers. Moreover, these high-resolution approaches promise to improve tumor classification and patient stratification, thereby improving the precision of future clinical trials.

Application of scRNA-seq in biomarker discovery

Using an scRNA-seq approach, a study found that HCC expressing the predictive markers FKBP10, ATP1A2, NT5DC2, UGT3A2, PYCR1, CKB, GPX7, DNMT3B, GSTP1, and OXCT1 activate a metabolic transcriptional program that influences epigenetic regulation.⁶⁰ In another study, an scRNA-seq approach was used to calculate a metabolic score predicting metastatic potential in HCC based on the combined tumor expression of DNMT3B and PFKFB4. Emerging evidence suggests that circular RNAs (circRNAs) and lncRNAs contribute substantially to the molecular etiology of HCC.⁶¹ circRNAs and lncRNAs are hypothesized to play important roles in the etiology of HCC. Kaplan- Meier plot analysis showed that of the 55 differentially expressed mRNAs in the circRNA/lncRNA-miRNA-mRNA network, the following genes were most strongly associated with the prognosis of patients with HCC: CPEB3, EFNB3, FATA4, GH receptor, GSTZ1, KLF8, MFAP4, PAIP2B, PHACTR3, PITPNM3, RPS6KA6, RSPO3, SLITRK6, SMOC1, STEAP4, SYT1, TMEM132E, TSPAN11, and ZFPM2.⁶²

APPLICATION OF scRNA-SEQ IN LIQUID BIOPSY AND NONINVASIVE MONITORING

Liquid biopsy is a minimally invasive method for detecting tumor-derived materials in blood. It enables the acquisition of genetic, epigenetic, transcriptomic, and proteomic data. This method is especially important in HCC because biopsies are rarely performed to diagnose HCC owing to the widespread reliance on imaging techniques such as computed tomography (CT) or magnetic resonance imaging (MRI) scans. Tumor-derived materials for liquid biopsy include cell-free DNA, circulating tumor cells (CTCs), and extracellular vesicles. The presence of CTCs is regarded as a valuable biomarker for the diagnosis and prognosis of HCC.

scRNA-seq is a promising approach for analyzing gene expression in individual CTCs. However, these cells are present in extremely low numbers (approximately 1-10 per million white blood cells [WBCs]). Therefore, their enrichment is critical before scRNA-seq and is typically achieved using various immunoaffinity-based or physical property-based methods. The Cell-Search^® System (Menarini Silicon Biosystems Inc, Huntington Valley, PA, USA), an epithelial cell adhesion molecule (EpCAM)-based immunoaffinity platform, is currently the only United States Food and Drug Administration (FDA)-approved method for CTC detection. It employs antibodies targeting EpCAMs to isolate and enrich CTCs in metastatic breast, colorectal, and prostate cancers. This system captures CTCs using ferromagnetic beads coated with anti-EpCAM antibodies and isolates them via magnetically activated cell sorting. However, CTCs commonly exhibit epithelial-to-mesenchymal transition (EMT), and in the context of HCC, they arise from hepatocytes rather than from classical epithelial cell lineages. Therefore, the use of anti-EpCAM antibodies alone may not be sufficient for effective enrichment of HCC CTCs. To overcome this, newer strategies have incorporated multiple antibodies to improve the capture efficiency. Advanced microfluidic platforms, such as CTC-Chip^TM (Johnson & Johnson, New Brunswick, NJ, USA) and NanoVelcro Chip^TM (University of California, Los Angeles, Los Angeles, CA, USA), have been developed to increase the frequency and strength of interactions between CTCs and functionalized chip surfaces, resulting in markedly improved capture efficiency. Alternatively, physical property-based approaches utilize techniques, such as microfiltration, density gradient centrifugation, dielectrophoresis, and microfluidic systems, to isolate CTCs. All of these methods are antigen-independent; therefore, they can address the issue of EpCAM loss during the EMT of CTCs. Additionally, CTCs do not need to be separated from the coated antibodies, which makes these methods cost-effective. The Parsortix^® PC1 System (CelLBxHealth, Guildford, UK) represents a microfluidic technology recently cleared by the FDA that isolates CTCs from patients with metastatic breast cancer by exploiting differences in cell size and deformability. However, even with these technologies, more than 1,000 WBCs per CTC often remain in the background. Therefore, immunofluorescence staining is commonly used to distinguish CTCs from the background WBCs. Alternative strategies have applied PCR-based assays to characterize CTC expression signatures through the analysis of a restricted set of genes, as shown in recent studies on HCC.⁶³ However, comprehensive whole-genome transcriptomic profiling of CTCs from patients with HCC remains scarce.

Advances in single-cell technology offer a feasible path for obtaining such data, providing a foundation for the development of more precise biomarkers for HCC. However, only a few studies have employed scRNA-seq to detect and characterize CTCs in HCC. D’Avola et al.⁶⁴ reported a sequential approach combining imaging flow cytometry with high-density scRNA-seq to detect and characterize CTCs in patients with HCC. Genome-wide expression profiling of CTCs using this approach demonstrated that the presence of HCC upregulated lncRNA (HULC), SPINK1, IGF2, SPP1, and BRIC5 in the CTCs of the hepatic lineage. Moreover, the absence of circulating hepatic lineage cells in the blood of healthy individuals and patients with non-malignant chronic liver diseases further supports the malignant origin of CTCs. In addition to the aforementioned genes, multiple other genes exhibited differential expression between the two candidate CTCs detected in patient 1, underscoring the transcriptomic heterogeneity within the CTC compartment. This study suggests that high-density expression profiling using scRNA-seq helps trace cancer heterogeneity in CTCs.⁶⁴ Building on this concept, researchers performed single-cell transcriptomic analysis to identify cell type-specific RNA markers in plasma samples from patients with HCC. In this study, scRNA-seq identified six major cell clusters in HCC tissues compared with paired non-HCC tissues. These profiles were then used to derive cell-type-specific gene signature (CELSIG) scores from plasma RNA. Notably, the hepatocyte-like CELSIG score was significantly elevated in the preoperative plasma of patients with HCC compared with that of patients without HCC. These findings highlight the use of combining scRNA-seq with plasma RNA analysis to develop noninvasive and dynamic biomarkers for cancer detection and monitoring.⁶⁵ Collectively, these studies underscore the significance of scRNA-seq as an emerging approach for biomarker identification in HCC using liquid biopsy. Various applications of scRNA-seq in cancer are shown in Fig. 4.

CHALLENGES AND PERSPECTIVES

Although scRNA-seq has transformed our understanding of organ complexity, it faces several challenges: tissue dissociation can alter gene expression and must be finely tuned, the approach remains costly and requires extensive bioinformatics expertise, and spatial transcriptomics, while promising, still requires validation in liver studies. Emerging single-cell multiomics methods will further enrich our insights by integrating protein, RNA, and DNA analyses; however, they must become more affordable.

Finally, the development of reliable histological or imaging biomarkers could help predict treatment responses and patient outcomes without the need for full single-cell profiling. Nevertheless, integrating single-cell and multi-omics analyses into the real-world diagnostics of HCC is challenging. Most clinical specimens are limited to small biopsies or formalin-fixed paraffin-embedded (FFPE) tissues, where RNA degradation and rapid tissue handling pipelines hinder high-quality single-cell isolation and sequencing.^66,67 Furthermore, these samples may not adequately capture the spatial and cellular heterogeneity of HCC, leading to sampling bias. New methods, including single-nucleus RNA sequencing^68,69 and FFPE-compatible spatial transcriptomics,^70,71 address these shortcomings; however, they are costly and technically challenging for routine use in clinical practice.

CONCLUSION

scRNA-seq has transformed our ability to dissect liver biology by resolving the diversity of cell types and states and mapping intercellular communication in health and disease. Selecting an appropriate scRNA-seq protocol requires balancing experimental goals, desired gene coverage, and budget constraints. Advances in computational analysis, spatial transcriptomics, and hybrid workflows that combine scRNA-seq with spatial data now enable us to study gene expression within the intact, architecturally complex liver. This approach has yielded key insights into the hepatic zonation, mechanisms of regeneration, and cellular ecosystems that drive chronic liver disease and cancer. Because liver pathology involves intricate networks of multiple cell populations, scRNAseq offers an unprecedented window into these microenvironments. The current challenge lies in converting these molecular and cellular discoveries into targeted therapies to address the pressing clinical needs of hepatology.

Article information

Conflicts of Interest

The authors have no conflicts of interests to declare.

Ethics Statement

This review article is fully based on articles which have already been published and did not involve additional patient participants. Therefore, IRB approval is not necessary.

Funding Statement

Not applicable.

Data Availability

Not applicable.

Author Contributions

Conceptualization: CJ, RG

Data curation: CJ, SP

Formal analysis: CJ, SP

Investigation: CJ, MP

Methodology: CJ, SP, MP

Project administration: CJ

Resources: RG

Supervision: RG

Writing - original draft: SP, RG

Writing - review & editing: RG

Figure 1.

Tumor-causing agents and the ecosystem of primary liver cancer. Multiple tumor-causing agents, including toxins, diabetes, hepatitis B virus (HBV), hepatitis C virus (HCV), alcohol, non-alcoholic fatty liver disease (NAFLD), and gene mutations, contribute to hepatic carcinogenesis. These factors induce genetic and cellular alterations in hepatocytes, resulting in intertumoral heterogeneity (variations between patients) and intratumoral heterogeneity (variations within the same tumor). The heterogeneous tumor microenvironment consists of different cell types, including hepatocytes, hepatoma cells, T cells, B cells, macrophages, stromal cells, dendritic cells, and vasculature. The diagram also illustrates possible sites of metastasis (brain, bone, lung, and kidney).

Figure 2.

Workflow for single-cell RNA (scRNA) isolation, sequencing, and data analysis. The schematic illustrates the entire procedure of scRNA sequencing, where the sample is collected and stored before it can be isolated using various methods, such as fluorescence-activated cell sorting (FACS), magnetic-activated cell sorting (MACS), laser capture microdissection (LCM), and microfluidic-based separation. Following isolation, single cells undergo lysis, reverse transcription, and library preparation for sequencing. Various sequencing platforms are used in the sequencing, including single-cell tagged reverse transcription sequencing (STRT-seq), switching mechanism at the 5’ end of RNA template sequencing (SMART-seq), cell expression by linear amplification and sequencing (CEL-seq), droplet-based sequencing (Drop-seq), or 10x Chromium Genomics. Dimensionality-reduction algorithms, such as t-distributed stochastic neighbor embedding (t-SNE), are used to visualize the resulting transcriptomic data.

Figure 3.

Please check copyright of figure. Importance of single-cell multi-omics in understanding tumor heterogeneity. This figure depicts the role of single-cell multi-omics in cancer diagnosis and analysis of intertumoral and intratumoral heterogeneity. Multi-omics methods can provide a complete picture of tumor biology by combining genomics, transcriptomics, proteomics, and metabolomics. These analyses reveal the main molecular signatures and pathways, including TP53, RB1, CTNNB1, PROM1, and CD47, as well as changes in various cell types, such as cancer stem cells, cancer-associated fibroblasts (CAFs), tumor-associated macrophages (TAMs), T lymphocytes, and Kupffer cells. Dysregulated pathways such as glycolysis, hypoxia response, immune evasion, angiogenesis, and metastasis are mentioned as the drivers of tumor progression. Single-cell multi-omics insights aid in the development of personalized medicine, drug targets, biomarker discovery, patient stratification, and clinical translation through multi-phase therapeutic trials. TP53, tumor protein p53; RB1, retinoblastoma gene; CTNNB1, β-catenin gene; PROM1, prominin-1 (CD133); CD47, cluster of differentiation 47; CSC, cancer stem cells; MAC, macrophages.

Figure 4.

Applications of single-cell sequencing. This figure demonstrates the uses of single-cell RNA sequencing (scRNA-seq) in cancer. The schematic shows the biomarker discovery through predictive markers that trigger metabolic transcriptional programs and affect epigenetic regulation. It also describes the ability to predict metastatic potential and identify prognostic markers with the help of the Kaplan-Meier analysis of circular RNA (circRNA) and long non-coding RNA (lncRNA) profiles. The noninvasive tumor monitoring methods based on liquid biopsy are circulating tumor cells (CTCs), cell-free DNA (cfDNA), and extracellular vesicles, along with single-cell isolation. Multiplex protein analysis, single-nucleotide polymorphism (SNP) genotyping, chromatin immunoprecipitation (CHIP) assays, DNA methylation, RNA, and nucleosome profiling are examples of integrative single-cell assays that contribute to the comprehensive diagnosis and prognosis of cancer.

References

• 1. Peeters F, Cappuyns S, Piqué-Gili M, Phillips G, Verslype C, Lambrechts D, et al. Applications of single-cell multi-omics in liver cancer. JHEP Rep 2024;6:101094. ArticlePubMedPMC
• 2. Wang M, Ye Q, Mao D, Li H. Research progress in liver-regenerating microenvironment and DNA methylation in hepatocellular carcinoma: the role of traditional Chinese medicine. Med Sci Monit 2020;26:e920310.ArticlePubMedPMC
• 3. Wang Y, Wang P, Zhang Z, Zhou J, Fan J, Sun Y. Dissecting the tumor ecosystem of liver cancers in the single-cell era. Hepatol Commun 2023;7:e0248.ArticlePubMedPMC
• 4. Ye J, Lin Y, Liao Z, Gao X, Lu C, Lu L, et al. Single cell-spatial transcriptomics and bulk multi-omics analysis of heterogeneity and ecosystems in hepatocellular carcinoma. NPJ Precis Oncol 2024;8:262. ArticlePubMedPMCPDF
• 5. Safri F, Nguyen R, Zerehpooshnesfchi S, George J, Qiao L. Heterogeneity of hepatocellular carcinoma: from mechanisms to clinical implications. Cancer Gene Ther 2024;31:1105−1112.ArticlePubMedPMCPDF
• 6. Li L, Wang H. Heterogeneity of liver cancer and personalized therapy. Cancer Lett 2016;379:191−197.ArticlePubMed
• 7. Wu L, Yan J, Bai Y, Chen F, Zou X, Xu J, et al. An invasive zone in human liver cancer identified by Stereo-seq promotes hepatocyte-tumor cell crosstalk, local immunosuppression and tumor progression. Cell Res 2023;33:585−603.ArticlePubMedPMCPDF
• 8. Cheng Y, Chen X, Feng L, Yang Z, Xiao L, Xiang B, et al. Stromal architecture and fibroblast subpopulations with opposing effects on outcomes in hepatocellular carcinoma. Cell Discov 2025;11:1. ArticlePubMedPMCPDF
• 9. Tian B, Li Q. Single-cell sequencing and its applications in liver cancer. Front Oncol 2022;12:857037. ArticlePubMedPMC
• 10. Saviano A, Henderson NC, Baumert TF. Single-cell genomics and spatial transcriptomics: discovery of novel cell states and cellular interactions in liver physiology and disease biology. J Hepatol 2020;73:1219−1230.ArticlePubMedPMC
• 11. Zhang Q, He Y, Luo N, Patel SJ, Han Y, Gao R, et al. Landscape and dynamics of single immune cells in hepatocellular carcinoma. Cell 2019;179:829−845.e20.ArticlePubMedPMC
• 12. Dong X, Wang F, Liu C, Ling J, Jia X, Shen F, et al. Single-cell analysis reveals the intra-tumor heterogeneity and identifies MLXIPL as a biomarker in the cellular trajectory of hepatocellular carcinoma. Cell Death Discov 2021;7:14. ArticlePubMedPMCPDF
• 13. Ma L, Heinrich S, Wang L, Keggenhoff FL, Khatib S, Forgues M, et al. Multiregional single-cell dissection of tumor and immune cells reveals stable lock-and-key features in liver cancer. Nat Commun 2022;13:7533. ArticlePubMedPMCPDF
• 14. Adiconis X, Haber AL, Simmons SK, Moonshine AL, Ji Z, Busby MA, et al. Comprehensive comparative analysis of 5'-end RNA-sequencing methods. Nat Methods 2018;15:505−511.ArticlePubMedPMCPDF
• 15. Chen J, Suo S, Tam PP, Han JJ, Peng G, Jing N. Spatial transcriptomic analysis of cryosectioned tissue samples with Geo-seq. Nat Protoc 2017;12:566−580.ArticlePubMedPDF
• 16. Danielski K. Guidance on Processing the 10x genomics single cell gene expression assay. Methods Mol Biol 2023;2584:1−28.ArticlePubMed
• 17. Hashimshony T, Wagner F, Sher N, Yanai I. CEL-Seq: single-cell RNA-Seq by multiplexed linear amplification. Cell Rep 2012;2:666−673.ArticlePubMedPMC
• 18. Huang XT, Li X, Qin PZ, Zhu Y, Xu SN, Chen JP. Technical advances in single-cell RNA sequencing and applications in normal and malignant hematopoiesis. Front Oncol 2018;8:582. ArticlePubMedPMC
• 19. Klein AM, Macosko E. InDrops and Drop-seq technologies for single-cell sequencing. Lab Chip 2017;17:2540−2541.ArticlePubMed
• 20. Natarajan KN. Single-cell tagged reverse transcription (STRT-seq). Methods Mol Biol 2019;1979:133−153.ArticlePubMed
• 21. Picelli S, Faridani OR, Björklund AK, Winberg G, Sagasser S, Sandberg R. Full-length RNA-seq from single cells using Smart-seq2. Nat Protoc 2014;9:171−181.ArticlePubMedPMCPDF
• 22. Sheng K, Cao W, Niu Y, Deng Q, Zong C. Effective detection of variation in single-cell transcriptomes using MATQ-seq. Nat Methods 2017;14:267−270.ArticlePubMedPDF
• 23. Luo C, Rivkin A, Zhou J, Sandoval JP, Kurihara L, Lucero J, et al. Robust single-cell DNA methylome profiling with snmC-seq2. Nat Commun 2018;9:3824. ArticlePubMedPMCPDF
• 24. Ma S, Zhang Y. Profiling chromatin regulatory landscape: insights into the development of ChIP-seq and ATAC-seq. Mol Biomed 2020;1:9. ArticlePubMedPMCPDF
• 25. Mulqueen RM, Pokholok D, O'Connell BL, Thornton CA, Zhang F, O'Roak BJ, et al. High-content single-cell combinatorial indexing. Nat Biotechnol 2021;39:1574−1580.ArticlePubMedPMCPDF
• 26. Nakabayashi K, Yamamura M, Haseagawa K, Hata K. Reduced representation bisulfite sequencing (RRBS). Methods Mol Biol 2023;2577:39−51.ArticlePubMed
• 27. Nakato R, Sakata T. Methods for ChIP-seq analysis: a practical workflow and advanced applications. Methods 2021;187:44−53.ArticlePubMed
• 28. Song L, Crawford GE. DNase-seq: a high-resolution technique for mapping active gene regulatory elements across the genome from mammalian cells. Cold Spring Harb Protoc 2010;2010:pdb.prot5384. ArticlePubMedPMC
• 29. Wang K, Xiao Z, Yan Y, Ye R, Hu M, Bai S, et al. Simple oligonucleotide-based multiplexing of single-cell chromatin accessibility. Mol Cell 2021;81:4319−4332.e10.ArticlePubMedPMC
• 30. Nouri N, Kurlovs AH, Gaglia G, de Rinaldis E, Savova V. Scaling up single-cell RNA-seq data analysis with CellBridge workflow. Bioinformatics 2023;39:btad760. ArticlePubMedPMCPDF
• 31. He J, Lin L, Chen J. Practical bioinformatics pipelines for single-cell RNA-seq data analysis. Biophys Rep 2022;8:158−169.PubMedPMC
• 32. Danese A, Richter ML, Chaichoompu K, Fischer DS, Theis FJ, Colomé-Tatché M. EpiScanpy: integrated single-cell epigenomic analysis. Nat Commun 2021;12:5228. ArticlePubMedPMCPDF
• 33. Cappuyns S, Philips G, Vandecaveye V, Boeckx B, Schepers R, Van Brussel T, et al. PD-1- CD45RA+ effector-memory CD8 T cells and CXCL10+ macrophages are associated with response to atezolizumab plus bevacizumab in advanced hepatocellular carcinoma. Nat Commun 2023;14:7825. ArticlePubMedPMCPDF
• 34. Song HW, Martin J, Shi X, Tyznik AJ. Key considerations on CITE-seq for single-cell multiomics. Proteomics 2025;25:206−213.ArticlePubMedPMC
• 35. Li W, Bazaz SR, Mayoh C, Salomon R. Analytical workflows for single-cell multiomic data using the BD rhapsody platform. Curr Protoc 2024;4:e963.ArticlePubMedPDF
• 36. Wang S, Xie J, Zou X, Pan T, Yu Q, Zhuang Z, et al. Single-cell multiomics reveals heterogeneous cell states linked to metastatic potential in liver cancer cell lines. iScience 2022;25:103857. ArticlePubMedPMC
• 37. Huo X, Hu S, Zhao C, Zhang Y. Dr.seq: a quality control and analysis pipeline for droplet sequencing. Bioinformatics 2016;32:2221−2223.ArticlePubMedPDF
• 38. Yu L, Wang X, Mu Q, Tam SST, Loi DSC, Chan AKY, et al. scONE-seq: a single-cell multi-omics method enables simultaneous dissection of phenotype and genotype heterogeneity from frozen tumors. Sci Adv 2023;9:eabp8901.ArticlePubMedPMC
• 39. Zachariadis V, Cheng H, Andrews N, Enge M. A highly scalable method for joint whole-genome sequencing and gene-expression profiling of single cells. Mol Cell 2020;80:541−553.e5.ArticlePubMed
• 40. Theunis K, Vanuytven S, Claes I, Geurts J, Rambow F, Brown D, et al. Single-cell genome and transcriptome sequencing without upfront whole-genome amplification reveals cell state plasticity of melanoma subclones. Nucleic Acids Res 2025;53:gkaf173. ArticlePubMedPMCPDF
• 41. Lu X, Zhu M, Pei X, Ma J, Wang R, Wang Y, et al. Super-enhancers in hepatocellular carcinoma: regulatory mechanism and therapeutic targets. Cancer Cell Int 2025;25:7. ArticlePubMedPMCPDF
• 42. Li J, Wang J, Wang Y, Zhao X, Su T. E2F1 combined with LINC01004 super-enhancer to promote hepatocellular carcinoma cell proliferation and metastasis. Clin Epigenetics 2023;15:17. ArticlePubMedPMCPDF
• 43. Hou Y, Guo H, Cao C, Li X, Hu B, Zhu P, et al. Single-cell triple omics sequencing reveals genetic, epigenetic, and transcriptomic heterogeneity in hepatocellular carcinomas. Cell Res 2016;26:304−319.ArticlePubMedPMCPDF
• 44. Plongthongkum N, Diep D, Chen S, Lake BB, Zhang K. Scalable dual-omics profiling with single-nucleus chromatin accessibility and mRNA expression sequencing 2 (SNARE-seq2). Nat Protoc 2021;16:4992−5029.ArticlePubMedPDF
• 45. Chen S, Lake BB, Zhang K. High-throughput sequencing of the transcriptome and chromatin accessibility in the same cell. Nat Biotechnol 2019;37:1452−1457.ArticlePubMedPMCPDF
• 46. Zhu C, Yu M, Huang H, Juric I, Abnousi A, Hu R, et al. An ultra high-throughput method for single-cell joint analysis of open chromatin and transcriptome. Nat Struct Mol Biol 2019;26:1063−1070.ArticlePubMedPMCPDF
• 47. Chen S, Huang C, Liao G, Sun H, Xie Y, Liao C, et al. Distinct single-cell immune ecosystems distinguish true and de novo HBV-related hepatocellular carcinoma recurrences. Gut 2023;72:1196−1210.ArticlePubMed
• 48. Du W, Huang Y, Chen X, Deng Y, Sun Y, Yang H, et al. Discovery of a PROTAC degrader for METTL3-METTL14 complex. Cell Chem Biol 2024;31:177−183.e17.ArticlePubMed
• 49. Wang H, Liang Y, Liu Z, Zhang R, Chao J, Wang M, et al. POSTN+ cancer-associated fibroblasts determine the efficacy of immunotherapy in hepatocellular carcinoma. J Immunother Cancer 2024;12:e008721.ArticlePubMedPMC
• 50. Jin H, Kim W, Yuan M, Li X, Yang H, Li M, et al. Identification of SPP1+ macrophages as an immune suppressor in hepatocellular carcinoma using single-cell and bulk transcriptomics. Front Immunol 2024;15:1446453. ArticlePubMedPMC
• 51. Zhu GQ, Tang Z, Huang R, Qu WF, Fang Y, Yang R, et al. CD36+ cancer-associated fibroblasts provide immunosuppressive microenvironment for hepatocellular carcinoma via secretion of macrophage migration inhibitory factor. Cell Discov 2023;9:25. ArticlePubMedPMCPDF
• 52. Liu Y, Dong G, Yu J, Liang P. Integration of single-cell and spatial transcriptomics reveals fibroblast subtypes in hepatocellular carcinoma: spatial distribution, differentiation trajectories, and therapeutic potential. J Transl Med 2025;23:198. ArticlePubMedPMCPDF
• 53. Park JG, Roh PR, Kang MW, Cho SW, Hwangbo S, Jung HD, et al. Intrahepatic IgA complex induces polarization of cancer-associated fibroblasts to matrix phenotypes in the tumor microenvironment of HCC. Hepatology 2024;80:1074−1086.PubMed
• 54. Alvarez M, Benhammou JN, Darci-Maher N, French SW, Han SB, Sinsheimer JS, et al. Human liver single nucleus and single cell RNA sequencing identify a hepatocellular carcinoma-associated cell-type affecting survival. Genome Med 2022;14:50. ArticlePubMedPMCPDF
• 55. Taniai T, Shimada S, Akiyama Y, Hatano M, Yasukawa K, Igarashi Y, et al. Integrative transcriptome profiling elucidates molecular and immunovascular characteristics of macrotrabecular HCC. Hepatology 2026;83:231−248.ArticlePubMedPMC
• 56. Zhou Q, Wu J, Bei J, Zhai Z, Chen X, Liang W, et al. Integration of single-cell sequencing and drug sensitivity profiling reveals an 11-gene prognostic model for liver cancer. Hum Genomics 2024;18:132. ArticlePubMedPMCPDF
• 57. Wu R, Guo W, Qiu X, Wang S, Sui C, Lian Q, et al. Comprehensive analysis of spatial architecture in primary liver cancer. Sci Adv 2021;7:eabg3750.ArticlePubMedPMC
• 58. Zhou PY, Zhou C, Gan W, Tang Z, Sun BY, Huang JL, et al. Single-cell and spatial architecture of primary liver cancer. Commun Biol 2023;6:1181. ArticlePubMedPMCPDF
• 59. Lehrich BM, Tao J, Liu S, Hirsch TZ, Yasaka TM, Cao C, et al. Development of mutated β-catenin gene signature to identify CTNNB1 mutations from whole and spatial transcriptomic data in patients with HCC. JHEP Rep 2024;6:101186. ArticlePubMedPMC
• 60. Monge C, Francés R, Marchio A, Pineau P, Desterke C, Mata-Garrido J. Characterization of an activated metabolic transcriptional program in hepatoblastoma tumor cells using scRNA-seq. Int J Mol Sci 2024;25:13044. ArticlePubMedPMC
• 61. Desterke C, Francés R, Monge C, Marchio A, Pineau P, Mata-Garrido J. Single-cell RNA-seq analysis links DNMT3B and PFKFB4 transcriptional profiles with metastatic traits in hepatoblastoma. Biomolecules 2024;14:1394. ArticlePubMedPMC
• 62. Zhou F, Kang Q, Ma J, Cai J, Chen Y, Qu K, et al. Integrated analysis of RNA-seq in hepatocellular carcinoma reveals competing endogenous RNA network composed of circRNA, lncRNA, and mRNA. Medicine (Baltimore) 2023;102:e32915.ArticlePubMedPMC
• 63. Park J, Lee YT, Agopian VG, Liu JS, Koltsova EK, You S, et al. Liquid biopsy in hepatocellular carcinoma: challenges, advances, and clinical implications. Clin Mol Hepatol 2025;31(Suppl): S255−S284.ArticlePubMedPMCPDF
• 64. D'Avola D, Villacorta-Martin C, Martins-Filho SN, Craig A, Labgaa I, von Felden J, et al. High-density single cell mRNA sequencing to characterize circulating tumor cells in hepatocellular carcinoma. Sci Rep 2018;8:11570. PubMedPMC
• 65. Vong JSL, Ji L, Heung MMS, Cheng SH, Wong J, Lai PBS, et al. Single cell and plasma RNA sequencing for RNA liquid biopsy for hepatocellular carcinoma. Clin Chem 2021;67:1492−1502.ArticlePubMedPDF
• 66. Levin Y, Talsania K, Tran B, Shetty J, Zhao Y, Mehta M. Optimization for sequencing and analysis of degraded FFPE-RNA samples. J Vis Exp 2020;160:e61060.ArticlePubMedPMC
• 67. González-Martínez S, Palacios J, Carretero-Barrio I, Lanza VF, Piqueras MGC, Caniego-Casas T, et al. Single-cell RNA sequencing on formalin-fixed and paraffin-embedded (FFPE) tissue identified multi-ciliary cells in breast cancer. Cells 2025;14:197. ArticlePubMedPMC
• 68. Xu Z, Zhang T, Chen H, Zhu Y, Lv Y, Zhang S, et al. High-throughput single nucleus total RNA sequencing of formalin-fixed paraffin-embedded tissues by snRandom-seq. Nat Commun 2023;14:2734. ArticlePubMedPMCPDF
• 69. Guo Y, Wang W, Ye K, He L, Ge Q, Huang Y, et al. Single-nucleus RNA-seq: open the era of great navigation for FFPE tissue. Int J Mol Sci 2023;24:13744. ArticlePubMedPMC
• 70. Villacampa EG, Larsson L, Mirzazadeh R, Kvastad L, Andersson A, Mollbrink A, et al. Genome-wide spatial expression profiling in formalin-fixed tissues. Cell Genom 2021;1:100065. ArticlePubMedPMC
• 71. Matsunaga H, Arikawa K, Yamazaki M, Wagatsuma R, Ide K, Samuel AZ, et al. Reproducible and sensitive micro-tissue RNA sequencing from formalin-fixed paraffin-embedded tissues for spatial gene expression analysis. Sci Rep 2022;12:19511. ArticlePubMedPMCPDF
• 72. Krjutškov K, Katayama S, Saare M, Vera-Rodriguez M, Lubenets D, Samuel K, et al. Single-cell transcriptome analysis of endometrial tissue. Hum Reprod 2016;31:844−853.ArticlePubMedPMC
• 73. Tian L, Su S, Dong X, Amann-Zalcenstein D, Biben C, Seidi A, et al. scPipe: a flexible R/Bioconductor preprocessing pipeline for single-cell RNA-sequencing data. PLoS Comput Biol 2018;14:e1006361.ArticlePubMedPMC
• 74. Ewels PA, Peltzer A, Fillinger S, Patel H, Alneberg J, Wilm A, et al. The nfcore framework for community-curated bioinformatics pipelines. Nat Biotechnol 2020;38:276−278.ArticlePubMedPMCPDF
• 75. Parekh S, Ziegenhain C, Vieth B, Enard W, Hellmann I. zUMIs - a fast and flexible pipeline to process RNA sequencing data with UMIs. Gigascience 2018;7:giy059. ArticlePubMedPMCPDF
• 76. Hwang B, Lee JH, Bang D. Single-cell RNA sequencing technologies and bioinformatics pipelines. Exp Mol Med 2018;50:1−14.ArticlePubMedPMCPDF
• 77. Su M, Pan T, Chen QZ, Zhou WW, Gong Y, Xu G, et al. Data analysis guidelines for single-cell RNA-seq in biomedical studies and clinical applications. Mil Med Res 2022;9:68. ArticlePubMedPMCPDF
• 78. Yuan Q, Duren Z. Integration of single-cell multi-omics data by regression analysis on unpaired observations. Genome Biol 2022;23:160. ArticlePubMedPMCPDF
• 79. Zheng GX, Terry JM, Belgrader P, Ryvkin P, Bent ZW, Wilson R, et al. Massively parallel digital transcriptional profiling of single cells. Nat Commun 2017;8:14049. PubMedPMC
• 80. Satija R, Farrell JA, Gennert D, Schier AF, Regev A. Spatial reconstruction of single-cell gene expression data. Nat Biotechnol 2015;33:495−502.ArticlePubMedPMCPDF
• 81. Gong W, Pan X, Xu D, Ji G, Wang Y, Tian Y, et al. Benchmarking DNA methylation analysis of 14 alignment algorithms for whole genome bisulfite sequencing in mammals. Comput Struct Biotechnol J 2022;20:4704−4716.ArticlePubMedPMC
• 82. Müller F, Scherer M, Assenov Y, Lutsik P, Walter J, Lengauer T, et al. Rn-Beads 2.0: comprehensive analysis of DNA methylation data. Genome Biol 2019;20:55. PubMedPMC
• 83. Casado-Pelaez M, Bueno-Costa A, Esteller M. Single cell cancer epigenetics. Trends Cancer 2022;8:820−838.ArticlePubMed
• 84. Cao B, Luo H, Luo T, Li N, Shao K, Wu K, et al. The performance of whole genome bisulfite sequencing on DNBSEQ-Tx platform examined by different library preparation strategies. Heliyon 2023;9:e16571.ArticlePubMedPMC
• 85. Li Q, Hermanson PJ, Springer NM. Detection of DNA methylation by whole-genome bisulfite sequencing. Methods Mol Biol 2018;1676:185−196.ArticlePubMed
• 86. Krueger F, Andrews SR. Bismark: a flexible aligner and methylation caller for Bisulfite-seq applications. Bioinformatics 2011;27:1571−1572.ArticlePubMedPMCPDF
• 87. Song Q, Decato B, Hong EE, Zhou M, Fang F, Qu J, et al. A reference methylome database and analysis pipeline to facilitate integrative and comparative epigenomics. PLoS One 2013;8:e81148.ArticlePubMedPMC
• 88. Broche J, Kungulovski G, Bashtrykov P, Rathert P, Jeltsch A. Genome-wide investigation of the dynamic changes of epigenome modifications after global DNA methylation editing. Nucleic Acids Res 2021;49:158−176.ArticlePubMedPMCPDF
• 89. Akalin A, Kormaksson M, Li S, Garrett-Bakelman FE, Figueroa ME, Melnick A, et al. methylKit: a comprehensive R package for the analysis of genome-wide DNA methylation profiles. Genome Biol 2012;13:R87. ArticlePubMedPMCPDF
• 90. Park Y, Wu H. Differential methylation analysis for BS-seq data under general experimental design. Bioinformatics 2016;32:1446−1453.ArticlePubMedPMCPDF
• 91. Han L, Wu HJ, Zhu H, Kim KY, Marjani SL, Riester M, et al. Bisulfite-independent analysis of CpG island methylation enables genome-scale stratification of single cells. Nucleic Acids Res 2017;45:e77.ArticlePubMedPMC
• 92. Lienhard M, Grimm C, Morkel M, Herwig R, Chavez L. MEDIPS: genome-wide differential coverage analysis of sequencing data derived from DNA enrichment experiments. Bioinformatics 2014;30:284−286.ArticlePubMedPMCPDF
• 93. Grandi FC, Modi H, Kampman L, Corces MR. Chromatin accessibility profiling by ATAC-seq. Nat Protoc 2022;17:1518−1552.ArticlePubMedPMCPDF
• 94. Patel H, Espinosa-Carrasco J, Langer B, Ewels P, nf-core bot, Garcia MU, et al. nf-core/atacseq: [2.1.2] - 2022-08-07 [Internet]. Genève (CH): Zenodo; [cited 2025 Oct 10]. Available from: https://doi.org/10.5281/zenodo.8222875Article
• 95. Hitz BC, Lee JW, Jolanki O, Kagda MS, Graham K, Sud P, et al. The ENCODE uniform analysis pipelines. bioRxiv 2023 Apr 6. doi: 10.1101/2023.04.04.535623. [Epub ahead of print].
• 96. Landt SG, Marinov GK, Kundaje A, Kheradpour P, Pauli F, Batzoglou S, et al. ChIP-seq guidelines and practices of the ENCODE and modENCODE consortia. Genome Res 2012;22:1813−1831.ArticlePubMedPMC
• 97. Moreno-Gonzalez M, Sierra I, Kind J. A hitchhiker's guide to single-cell epigenomics: methods and applications for cancer research. Int J Cancer 2026;158:291−304.ArticlePubMedPMC
• 98. Hu Y, Shen F, Yang X, Han T, Long Z, Wen J, et al. Single-cell sequencing technology applied to epigenetics for the study of tumor heterogeneity. Clin Epigenetics 2023;15:161. ArticlePubMedPMCPDF
• 99. Cittaro D, Lazarevic D, Tonon G, Giannese F. Analyzing genomic and epigenetic profiles in single cells by hybrid transposase (scGET-seq). STAR Protoc 2023;4:102176. ArticlePubMedPMC
• 100. Mölder F, Jablonski KP, Letcher B, Hall MB, van Dyken PC, Tomkins-Tinch CH, et al. Sustainable data analysis with Snakemake. F1000Res 2021;10:33. ArticlePubMedPMCPDF
• 101. Köster J, Rahmann S. Snakemake - a scalable bioinformatics workflow engine. Bioinformatics 2012;28:2520−2522.ArticlePubMedPMCPDF
• 102. Ludwig CH, Bintu L. Mapping chromatin modifications at the single cell level. Development 2019;146:dev170217. ArticlePubMedPMCPDF
• 103. Bartosovic M, Kabbe M, Castelo-Branco G. Single-cell CUT&Tag profiles histone modifications and transcription factors in complex tissues. Nat Biotechnol 2021;39:825−835.ArticlePubMedPMCPDF
• 104. Daneshpajouh H, Chen B, Shokraneh N, Masoumi S, Wiese KC, Libbrecht MW. Continuous chromatin state feature annotation of the human epigenome. Bioinformatics 2022;38:3029−3036.ArticlePubMedPMCPDF
• 105. Fang R, Preissl S, Li Y, Hou X, Lucero J, Wang X, et al. Comprehensive analysis of single cell ATAC-seq data with SnapATAC. Nat Commun 2021;12:1337. ArticlePubMedPMCPDF
• 106. Song L, Wang Y, Wang C, Yu Z, Wang L, He W, et al. Integration of bulk RNA and single-cell analyses reveal distinct expression patterns of anoikis-related genes and the immunosuppressive role of NQO1+ macrophages in hepatocellular carcinoma. FASEB J 2025;39:e70654.PubMed
• 107. Liu C, Li M, Liu L, Xu Q, Zheng L, Wu C, et al. TGF-β1 induces autophagy and mediates the effect on macrophages differentiation in primary liver cancer. Int Immunopharmacol 2025;157:114799. ArticlePubMed
• 108. Xu W, Weng J, Zhao Y, Xie P, Xu M, Liu S, et al. FMO2+ cancer-associated fibroblasts sensitize anti-PD-1 therapy in patients with hepatocellular carcinoma. J Immunother Cancer 2025;13:e011648.ArticlePubMedPMC
• 109. Pang W, Wang Y, Lu X, Li M, Long F, Chen S, et al. Integrated spatial and single cell transcriptomics identifies PRKDC as a dual prognostic biomarker and therapeutic target in hepatocellular carcinoma. Sci Rep 2025;15:14834. ArticlePubMedPMCPDF
• 110. Zhang J, Li Z, Zhang Q, Ma W, Fan W, Dong J, et al. LAMA4+ CD90+ eCAFs provide immunosuppressive microenvironment for liver cancer through induction of CD8⁺ T cell senescence. Cell Commun Signal 2025;23:203. ArticlePubMedPMCPDF
• 111. Lin P, Qin Q, Gan XY, Pang JS, Wen R, He Y, et al. Integrating single-cell and bulk RNA sequencing data to characterize the heterogeneity of glycan-lipid metabolism polarization in hepatocellular carcinoma. J Transl Med 2025;23:358. ArticlePubMedPMCPDF
• 112. Long F, Zhong W, Zhao F, Xu Y, Hu X, Jia G, et al. DAB2+ macrophages support FAP+ fibroblasts in shaping tumor barrier and inducing poor clinical outcomes in liver cancer. Theranostics 2024;14:4822−4843.ArticlePubMedPMC
• 113. Lee SK, Lim J, Jhun JY, Moon J, Kim HS, Choi JY, et al. Landscape of T-cell exhaustion heterogeneity and HBV integration in virus-related HCC revealed by whole-exome, transcriptome, and single-cell sequencing. JHEP Rep 2025;7:101518. ArticlePubMedPMC
• 114. Ma M, Jin C, Dong Q. Intratumoral heterogeneity and immune microenvironment in hepatoblastoma revealed by single-cell RNA sequencing. J Cell Mol Med 2025;29:e70482.PubMedPMC
• 115. Guo DZ, Zhang X, Zhang SQ, Zhang SY, Zhang XY, Yan JY, et al. Single-cell tumor heterogeneity landscape of hepatocellular carcinoma: unraveling the pro-metastatic subtype and its interaction loop with fibroblasts. Mol Cancer 2024;23:157. ArticlePubMedPMCPDF
• 116. Wang C, Huang W, Zhong Y, Zou X, Liu S, Li J, et al. Single-cell multi-modal chromatin profiles revealing epigenetic regulations of cells in hepatocellular carcinoma. Clin Transl Med 2024;14:e70000.ArticlePubMedPMC
• 117. Liu C, Wang K, Mei J, Zhao R, Shen J, Zhang W, et al. Integrative single-cell and spatial transcriptome analysis reveals heterogeneity of human liver progenitor cells. Hepatol Commun 2025;9:e0662.ArticlePubMedPMC
• 118. Chen Y, Deng X, Li Y, Han Y, Peng Y, Wu W, et al. Comprehensive molecular classification predicted microenvironment profiles and therapy response for HCC. Hepatology 2024;80:536−551.ArticlePubMedPMC
• 119. Sun J, Zhang S, Liu Y, Liu K, Gu X. Exploring tumor endothelial cells heterogeneity in hepatocellular carcinoma: insights from single-cell sequencing and pseudotime analysis. PeerJ 2024;12:e18362.ArticlePubMedPMCPDF
• 120. Cui J, Zeng F, Tang M, Yin S. Single-cell sequencing reveals cell heterogeneity and aberrantly activated pathways associated with microvascular invasion in hepatocellular carcinoma. Front Cell Dev Biol 2025;13:1449624. ArticlePubMedPMC
• 121. He X, Liu F, Gong Q. Identification of a senescence-related transcriptional signature to uncover molecular subtypes and key genes in hepatocellular carcinoma. PLoS One 2024;19:e0311696.ArticlePubMedPMC
• 122. Wu WJ, Wang J, Chen F, Wang X, Lan B, Fu R, et al. Exploration of heterogeneity and recurrence signatures in hepatocellular carcinoma. Mol Oncol 2025;19:2388−2411.ArticlePubMedPMCPDF
• 123. Zhu XY, Liu WT, Hou XJ, Zong C, Yu W, Shen ZM, et al. CD34+CLDN5+ tumor associated senescent endothelial cells through IGF2-IGF2R signaling increased cholangiocellular phenotype in hepatocellular carcinoma. J Adv Res 2025;76:511−528.ArticlePubMedPMC
• 124. Li S, Lin Y, Gao X, Zeng D, Cen W, Su Y, et al. Integrative multi-omics analysis reveals a novel subtype of hepatocellular carcinoma with biological and clinical relevance. Front Immunol 2024;15:1517312. ArticlePubMedPMC

Figure & Data

References

Citations

Citations to this article as recorded by  

• The lncRNA–DNA Methylation Axis in Hepatocellular Carcinoma: Mechanisms, Epigenetic Plasticity, and Biological Implications
  Lingke Meng, Lingzhu Cheng, Yuanyuan Li, Yushan Guo, Na Li
  Biology.2026; 15(6): 458.     CrossRef